sign in sign up
<<
Home , European land mammal age

First complete heterosoricine shrew: A new and species from the of China

GERHARD STORCH and ZHUDING QIU

Storch, G. and Qiu, Z. 2004. First complete heterosoricine shrew: A new genus and species from the Miocene of China. Acta Palaeontologica Polonica 49 (3): 357–363.

Heterosoricinae are the oldest known soricids, their records dating back to the middle of North America and earli− est of Eurasia. They became extinct during the Miocene and were thus far only known from dental and cranial remains. For the first time, a virtually complete heterosoricine is described, coming from the early/ local− ity of Shanwang, Shandong Province, which is famous for the diversity and excellent preservation of its . Lusorex taishanensis gen. et sp. nov. is closely related to Wilsonosorex from the early Hemingfordian of North America. Both are unusual in sharing well−developed conules on the upper molars and reduced ectocingulids on the lowers, and most likely these sister taxa reflect faunal exchange between North America and NE Asia in early Miocene time. L. taishanensis was the size of a European common shrew, Sorex araneus. The heavy masticatory apparatus of the new heterosoricine con− trasts with its slender postcranial skeleton. Adaptively, L. taishanensis appears to be similar to the North American Blarina brevicauda in its strong masticatory apparatus, very short tail, and slight limb specializations toward fossorial habits. It differs from other soricids as far as is known by unfused tibia and fibula.

Key words: Soricidae, Heterosoricinae, Miocene, China, skeleton, adaptations, new taxon.

Gerhard Storch [[email protected]], Forschungsinstitut Senckenberg, Senckenberganlage 25, D−60325 Frankfurt am Main, Germany; Zhuding Qiu [[email protected]], Institute of Vertebrate and Paleoanthropology (IVPP), Academia Sinica, P.O. Box 643, Beijing 100044, China.

Introduction dentition. It represents the only complete heterosoricine known to date and provides for the first time information Shanwang, about 22 km northeast of the county seat Linqu in about the postcranial skeleton in this subfamily. In addition Shandong Province, China, is a former lake of early–middle to teeth, only lower jaws and rostral skull fragments are Miocene age. The lake deposits are 120 m thick and consist known from previously described heterosoricines. mainly of shale and diatomite, lying on basalt and overlain The assignment of the new taxon to heterosoricines is by basalt (Yan et al. 1983). The site is one of a few classical based on a suite of mandibular and dental characters: The localities in China and has a great reputation for the horizontal ramus of the mandible is deep and massive, the high taxonomic diversity and extraordinary preservation of temporal fossa is not pocketed, and the well excavated its fossils, including plants and animals. Several dozens of masseteric fossa is subdivided by a horizontal crest. Upper complete and articulated skeletons of have been and lower molars have a compact form, and P4 and M1–2 have no emargination of the distal outline and thus no poste− recovered since the 1930s. Five species of small mammals rior projection of the hypoconal flange. and thirteen species of larger mammals have been described thus far (Young 1937, 1977; Wang 1965; Li 1974; Zhang 1974; Xie 1979; Qiu 1981; Yan et al. 1983; Qiu et al. 1985, 1986; Qiu and Sun 1988). Biochronologically, the Shanwang Systematic paleontology Local Fauna is referred to the late Shanwangian Chinese Land Age and can be correlated with the late Order Soricomorpha Gregory, 1910 European Land Mammal Age and the European Family Soricidae Fischer von Waldheim, 1817 Land Mammal Zone MN 5 (Li et al. 1984; Qiu and Qiu Subfamily Heterosoricinae Viret and Zapfe, 1951 1995). This correlation is based mainly on the co−occurrence of , , Plesiaceratherium, and Lago− Genus Lusorex nov. meryx with rather complicated antlers, the latter first appea− Figs. 1–6. ring in Zone MN5 (Mein 1999). Type species: Lusorex taishanensis sp. nov. This paper describes a new heterosoricine shrew from Etymology: After Lu, Chinese abbreviated form of Shandong Province, Shanwang on the basis of a virtually complete skeleton and the provenance of the , and sorex (Latin) for shrew.

Acta Palaeontol. Pol. 49 (3): 357–363, 2004 http://app.pan.pl/acta49/app49−357.pdf 358 ACTA PALAEONTOLOGICA POLONICA 49 (3), 2004

Fig. 1. Lusorex taishanensis (holotype), early–middle Miocene of Shanwang, Shandong Province, China (IVPP V 13915). The soft body “shadow” is pre− served in black.

Diagnosis.—First and second upper and lower molars re− Measurements semble those of Wilsonosorex Martin, 1978. Differs from Tooth measurements in mm (P4–M3 greatest length × great− Wilsonosorex by its undivided mesostyle (well−divided in est width, other positions greatest length only; A = ante− Wilsonosorex), crest−like and Y−shaped metaconule (a high molars between 1st incisor and 4th premolar): pointed cusp in Wilsonosorex), and weaker paraconule of Upper teeth.— A1 1.24; A2 0.40; A3 0.60; A4 0.28; A5 0.80; M1–2 (strong and doubled in Wilsonosorex); discontinuous P4 1.96×1.68; M1 1.68×1.92; M2 1.44×1.68; M3 0.84×1.28. ectocingulid below protoconid and hypoconid (below proto− conid only in Wilsonosorex), and very sharp, laterally com− Lower teeth.—i1 3.36; a1 0.88; a2 0.52; a3 0.64; a4 0.44; pressed entoconid of m1–2 (less crest−like in Wilsonosorex). a5 0.88; m1 1.84; m2 1.60; m3 1.28. Small size as in W. bateslandensis. Additional diagnostic Skeletal measurements in mm (greatest length); in paren− features include rather small first upper and lower incisors. theses measurements of an adult specimen of extant Sorex Differs from all other heterosoricine genera in having well− araneus: developed conules on upper molares and reduced ectocin− Skull and body (premaxilla to pubis) 63.0 (62.4); skull gulids on lowers. Postcranial skeleton essentially soricid− 19.2 (19.7); tail ~24.5 (47.7); humerus 8.0 (7.4); ulna 10.5 like. Differs from other subfamilies, as far as known, by (10.5); radius 7.5 (7.8); hand 6.9 (7.3); pelvis 11.5 (11.8); unfused tibia and fibula. ilium 6.0 (5.5); femur 8.5 (7.8); tibia 11.3 (13.3); fibula 9.7 (fused to tibia over most of its length); hind foot 10.0 (11.8). Indices of limb segments; in parentheses Sorex araneus: Lusorex taishanensis sp. nov. crural index (tibia:femur × 100) 133 (170); intermembral in− Holotype: A virtually complete skeleton with preserved soft body out− dex (humerus + radius : femur + tibia × 100) 78 (72). line. Dentition of right side completely preserved. Collection of the In− stitute of Vertebrate Paleontology and Paleoanthropology (IVPP), Chi− Description nese Academy of Sciences, Beijing, V 13915. Dentition.—The teeth are heavily pigmented. The dental for− Type locality and horizon: Shanwang (Diatomite quarry), Linqu County, mula is 1−6−3/1−5−3. Shandong Province, China. Early–midddle Miocene (Late Shanwangian, The first upper incisor is rather small. The medial tine is MN 5 equivalent). broken off yet there is not much doubt that the tooth was Etymology: Taishanensis, in allusion to the Taishan Mountain, one of fissident. The broad apex is strongly curved ventrally. the five celebrated mountains of China, which is located close to the fos− There are five crowded, single−rooted upper antemolars sil site. between the first incisor and P4. The roots incline progres− Diagnosis.—As for genus. sively toward the vertical from A1 to A5. A1 is distinctly STORCH AND QIU—NEW HETEROSORICINE FROM MIOCENE OF CHINA 359

Fig. 2. Dentition of Lusorex taishanensis: photo− graph (A) and explanatory drawing of the same (B). Above in each: 1st upper incisor to 5th ante− molar in lingual view; middle: 1st lower incisor to 3rd molar in lingual view; below: 1st lower incisor and 2nd molar in labial view. larger than A2–5. It has a pointed and high anterolabial cusp. A compressed posterior wall of the main cusp delimits the labial sharp ridge runs from its apex transversely to a lingual cingular tooth margin. A2, A3, and A5 exhibit an anterolabial cusp and cuspule and subdivides the occlusal surface into a shallow an− a flat distolingual heel. A5 has a sloping distolabial ridge. The terior ledge and an expanded posterior basin. The laterally minute peg−like A4 is squeezed between A3 and A5.

http://app.pan.pl/acta49/app49−357.pdf 360 ACTA PALAEONTOLOGICA POLONICA 49 (3), 2004

The upper molars decrease in size from M1 to M3. M1 and M2 are sub−rectangular in outline. The width of the ectoloph is subequal to the inner half of the teeth trans− versely. The cusps are high and sharp; the inner cusps are anteroposteriorly compressed into a wide V−shape. Proto− cone, paracone, and metacone are of subequal height; the hypocone is lower. The mesostyle is undivided, para− and metastyle are ridge−like, and the preparacrista of M1 is shorter than the postmetacrista. The hypocone is isolated an− teriorly from the postprotocrista by a deep valley. The para− conule is small and its anterior and posterior arms are short. The metaconule is ridge−like; the premetaconule crista termi− nates at the base of the metacone, and the shorter post− metaconule crista is directed posteriorly. Thus, the two meta− conule wings and the postprotocrista create a crest−like Y−shaped pattern. The prominent postcingulum closes the deep transverse posterior basin. The M2 is similar to M1, ex− cept for the subequal lengths of prepara− and postmetacrista. Fig. 3. Lusorex taishanensis. A.4th upper premolar to 3rd molar in occlusal The M3 is reduced distally. The premetacrista is continuous view. B.3rd to 5th lower antemolar (partially) and 1st to 3rd molar in occlusal with the postcingulum; distinct metacone, hypocone, and view. White line delimits antemolars that are concealed by the upper conules are lacking. The postprotocrista terminates post− antemolars. eriorly in the trigon basin. The first lower incisor, like the upper, is rather small for a P4 is sub−triangular, with the labial side being much lon− heterosoricine shrew. Its spatulate apex is gently curved up− ger than the lingual one. The paracone is high and pointed, wards, the cutting edge is smoothly bicuspate, and the and has two weak sloping anterior crests extending to the enamel is not wrinkled. The strong root exhibits a deep longi− precingulum. The posterior wing of the paracone is sharp and tudinal groove along the lingual side. separated by a small notch from the postcingulum. The The five antemolars between the first incisor and m1 are protocone is small and crest−like; its anterior and posterior very crowded. They are inclined anteriorly and single−rooted; arms are continuous with the pre− and postcingulum, respec− the roots incline progressively toward the horizontal from a5 tively. The parastyle is small. The posterior basin is deep. to a1. The crowns overlap each other widely. The a1 and a5 A thin ridge connects the protocone to the base of the para− (interpreted as p4) predominate in size, a1 being slightly larger cone. A hypocone is absent. than p4. The a1 has a high and sharp main cusp on the procc. coronoideus articularis angularis anterolabial corner and a broad, flat and posterolingually ex− panded heel. Weak lingual and posterior cingulids are visible. The a2 is small; its asymmetric chevron−shaped crown is com− posed of a low main cusp anteriorly and a slightly postero− lingually expanded heel. The a3 resembles a1 but is distinctly smaller and lower. The very small a4 is squeezed between a3 and p4; it is morphologically similar to a2. The p4 is longer than wide and has an ovate occlusal outline. The high main cusp has a flat posterior face with a distinct posterolingual sloping ridge; its apex is antero−posteriorly compressed. The tooth shows a flat and slightly sloping heel and a weak contin− uous cingulid along the lingual and distal margins. The root of p4 is very strong. The lower molars are graded in size. All cusps are very slender and piercing. The entoconid is sharp and laterally compressed. The metaconid and entoconid are of the same height and somewhat lower than the protoconid; the para− conid is lower than metaconid and entoconid, and the hypo− conid is lowest. The oblique cristid terminates anteriorly be− neath the protocristid notch. The postcristid joins the poste− rior wall of the entoconid directly, it is fused with the ento− I1 i1 horizontal ramus for. mandibulare conid and there is no postentoconid valley. There is a deep Fig. 4. X−ray image of dentition and lower jaws of Lusorex taishanensis. V−shaped talonid notch between metaconid and entoconid. STORCH AND QIU—NEW HETEROSORICINE FROM MIOCENE OF CHINA 361

pocketed. The masseteric fossa is well−excavated and subdi− vided by a horizontal crest. The condylar process exhibits two articular facets separated by a rather long and thin interarticular bridge. The mandibular foramen is wide and lo− cated in a depression below the crest−like ventral border of the temporal fossa. The vertebral column includes 7 cervical (most likely), 13 thoracic, 6 lumbar, probably 3 sacral, and probably 13 cau− dal vertebrae. The caudals are simple and slender. The clavi− cles are well−developed. The scapula shows a narrow outline, a high spine, and a long acromion, which is apparently bifur− cated. The pelvis has a straight ilium, a wide obturator fora− men, and a narrow process of the pubis, which originates from the ventro−caudal edge of the pelvis. We cannot totally diss− miss the possibility that the pubic spines formed a symphysis. It seems, however, more likely that a true pubic symphysis is Fig. 5. X−ray image of the forelimb of Lusorex taishanensis. absent. The pubic condition may compare to that of sorico− morphs such as the American shrew mole, Neurotrichus gib− bsi, where there is no true symphysis but a small tendinous tibial crest tibia fibula calcaneus band between the pubic spines (see Reed 1951). The humerus is gently bowed in anterior view. The well− developed teres tubercle and part of the pectoral ridge are dis− cernible on the right humerus. Both terminate almost halfway down the bone. The extension of the distal end of the humerus compares with that of extant shrews in general. Ulna and ra− dius are somewhat stronger than in Blarina. The olecranon is relatively long and slightly inclined anteriorly. The posterior crest (see Reed 1951, for definition) of the ulna is well−devel− oped, particularly opposite the semilunar notch and the coronoid process. The presence of a proximal ridge on the olecranon cannot be substantiated since that area is damaged on both ulnae. The radius is almost straight. Only the distal carpal row is exposed on the right hand. The trapezium is larg− est, followed in size by the hamatum. The distal phalanges of the hands are relatively robust, their dorsopalmar thickness be− ing greater than in Blarina. They are rather straight. The femur is straight and shows a well−developed third trochanter distal to the greater trochanter. The shank departs morphologically from that of other soricids by its unfused tibia and fibula. However, the fibula is tightly pressed against the femur metatarsus phalanges I–III caudal vertebra tibia along the distal half of the bones, and proximally they di− Fig. 6. X−ray image of the shank and hindfoot of Lusorex taishanensis. verge like in other shrews. Thus, the fused and unfused condi− tion are functionally most likely about the same. The falciform process of the tibia is well−developed and extends along the The cingulids are weak and discontinuous at the bases of the proximal third of the bone. Few details of the hind foot can be protoconid and hypoconid. The trigonid basin is open lin− substantiated. The tuber calcanei is well−developed, and the gually, and the deep hypoflexid opens at the level of the distal phalanges appear to be rather straight. ectocingulid. The talonid of m1 is wider than the trigonid. The m2 closely resembles m1 except for the subequal width of trigonid and talonid. The m3 is basically similar to m2, but Comparisons its talonid is narrower than the trigonid. Skeleton.—No skull details are accessible. The horizontal Martin (1978) described the new genus Wilsonosorex to− ramus of the lower jaw is deep and robust. The coronoid pro− gether with two new species from early Hemingfordian de− cess is strong and apically rather wide. X−ray photos reveal a posits on the Great Plains, USA. Both species, W. conulatus long styliform angular process that is characteristic of shrews from the Martin Canyon beds at Quarry A, Colorado, and W. in general. The temporal fossa is deeply excavated but not bateslandensis from the Flint Hill Local Fauna, South Da−

http://app.pan.pl/acta49/app49−357.pdf 362 ACTA PALAEONTOLOGICA POLONICA 49 (3), 2004 kota, were based on characters of the first and second upper roperi. P. roperi is similar to L. taishanensis in a combina− molars. First and second lower molars, found at the same lo− tion of characters such as: undivided mesostyle of M1–2, fu− calities, were associated by Martin (1978) only with reserva− sion of postcristid and entoconid of m1–2, well−developed tion with the uppers. The dentition of the new Chinese entoconid of m3, occlusal morphology of A1, and division of heterosoricine confirms this association now without doubt. the masseteric fossa on the lower jaw (see Engesser 1979: pl. Wilsonosorex is an unusual heterosoricine and Lusorex 8c; Wilson 1960: fig. 21d; Wilson 1963: fig. 1). P. roperi de− corresponds to most of its highly diagnostic characters (but parts widely, however, from L. taishanensis in features such see Diagnosis for differences), including: as: Absence of conules on M1–2, distinct hypocone on P4, – distinct paraconules and metaconules on M1–2; strong ectocingulids on m1–2, enlarged and slightly wrin− – protocone plus conules of M1–2 anteriorly compressed kled first lower incisor, trapezoidal outline of P4, and dental into a sharp V−pattern, with hypocone making a separate formula 1−5−3/1−4−3 (see Engesser 1979: pls. 6b, 8b, c; Wil− short V next to and twinned with the protocone; son 1960: figs. 14, 16; Wilson 1963: fig. 1). – ectoloph and inner half of M1–2 of subequal length; – cusps of upper and lower molars high and pointed; – ectocingulid of m1 discontinuous below protoconid (in Lusorex also below hypoconid); Discussion – postcristid of m1–2 confluent with the entoconid; – small size. L. taishanensis compares in size with the extant Europen com− We consider Wilsonosorex and Lusorex as closely related mon shrew (Sorex araneus) except for its distinctly shorter tail sister taxa, resulting from faunal exchanges between NE and shorter tibia and hind foot (see Measurements). The Asia and North America, most probably in early Miocene lengths of the segments of the hindlimbs of L. taishanensis time. In Tedford et al. (1987) the North American Quarry A show more isometry than in S. araneus. The higher crural in− and Flint Hill Local Faunas are considered early Heming− dex of S. araneus (see Measurements) suggests enhanced fordian and around 18–19 Ma. The European Neogene Land cursorial abilities, while the indices of the limb bones, the Mammal Zone MN 5, the presumable correlate of the late rather robust forearms and ungual phalanges of the hand, and Shanwangian Chinese Land Mammal Age, is later, around the very short and weak tail of L. taishanensis appear to indi− 15–17 Ma according to Steininger (1999). Plausible candi− cate more fossorial habits. Semiaquatic species such as the Eu− dates for the ancestry of these unique shrews are unknown, ropean water shrew, Neomys fodiens, are long−tailed and have however, from either North America or Asia. One of the a still higher crural index than S. araneus. most striking features of Wilsonosorex and Lusorex is the The very strong and heavy masticatory apparatus of L. presence of distinct conules. We consider this a plesiomor− taishanensis contrasts with its delicate postcranial skeleton phic character for heterosoricines but it is unknown from and it is also a striking difference between L. taishanensis heterosoricine taxa of earlier occurrences in North America, and S. araneus, where the lower jaws and the dentition are Asia, and Europe (and from the subfamily in general). To our much more slender. Among living shrews the North Ameri− knowledge upper molars of the genera Quercysorex Enges− can short−tailed shrew, Blarina brevicauda, also exhibits a ser, 1975, Eocene/Oligocene of Europe, Gobisorex Sulim− short tail, relatively strong forearms and hands, and a rela− ski, 1970, Oligocene/Miocene of eastern Asia, and Mongolo− tively heavy dentition, and it may represent adaptively an an− sorex Qiu, 1996, Miocene of eastern Asia, are unknown. The alogue of L. taishanensis. B. brevicauda is the most fossorial respective lowers, however, do not appear to be especially of American shrews and constructs surface and subsurface similar to the m1–2 of Wilsonosorex and Lusorex, e.g., they runways and subterranean burrows. Its diet consists of inver− exhibit a continuous and well−developed ectocingulid on m1 tebrates such as insects and snails, small vertebrates, and while it is weak and discontinuous in the latter genera. plant items. Martin and Green (1984) tentatively assigned two M1 The outline of the black soft body “shadow” is ovate. The and one m1 from the Hemingfordian age Black Bear Quarry II locality in the Rosebud Formation, South Dakota, to Wil− tail was obviously covered with short and sparse hairs. Ear sonosorex sp. They suggested that the specimens possibly pinnae and details of body hair cannot be substantiated. More represent a new heterosoricine genus. The Hemmingfordian preparation of the specimen will show whether gut contents Wilsonosorex sp. is clearly distinct from Lusorex in features are preserved. such as the well divided mesostyle and anteroposteriorly less compressed protocone and hypocone of M1, and the continu− ous ectocingulid on m1. Acknowledgements The only other taxon that shows some detailed resem− blance to L. taishanensis is Pseudotrimylus roperi (Wilson, We express our sincere thanks to Burkart Engesser and Lawrence 1960) from the early Hemingfordian of Quarry A, Colorado. J. Flynn for critical reading of the manuscript and helpful comments. We accept Gureev's original classification and confine the We thank Jörg Habersetzer for taking the x−ray photos. We are grateful genus Pseudotrimylus to include only the type species P. to L.J. Flynn for improvement of the English wording. STORCH AND QIU—NEW HETEROSORICINE FROM MIOCENE OF CHINA 363

on the newly found skeletons of Palaeomeryx from Shanwang, Shan− References dong. Vertebrata PalAsiatica 23: 187–203. Reed, C.A. 1951. Locomotion and appendicular anatomy in three soricoid Engesser, B. 1975. Revision der europäischen Heterosoricinae. Eclogae insectivores. The American Midland Naturalist 45 (3): 513–671. geologicae Helvetiae 68 (3): 649–671. Steininger, F.F. 1999. , geochronology and biochrono− Engesser, B. 1979. Relationships of some insectivores and rodents from the logy of the Miocene “European Land Mammal−Zones” (ELMMZ) and Miocene of North America and Europe. Bulletin of Carnegie Museum of the Miocene “Mammal−Zones (MN−Zones)”. In: G.E. Rössner, and K. Natural History 14: 1–68. Heissig (eds.), The Miocene Land Mammals of Europe, 9–24. Verlag Fischer von Waldheim, G. 1817. Soricidae. Mémoires de la Societé Imperiale Dr. Friedrich Pfeil, München. des Naturalistes du Moscou 5: 372. Sulimski A. 1970. On some Oligocene insectivore remains from Mongolia. Gregory, W.K. 1910. The orders of mammals. Bulletin of the American Mu− Palaeontologia Polonica 21: 53–70. seum of Natural History 27: 1–524. Tedford, R.H., Skinner, M.F., Fields, R.W., Rensberger, J.M., Whistler, Gureev A.A. 1971. Zemlerojki (Soricidae) fauny mira. 253 pp. Akademiâ D.P., Galusha, T., Taylor, B.E., Macdonald, J.R., and Webb, S.D. 1987. Nauk SSSR, Leningrad. Faunal succession and of the Arikareean through Hem− phillian interval (Late Oligocene through earliest epochs) in Li, C.−K. 1974. A probable geomyoid rodent from middle Miocene of North America. In: M.O. Woodburne (ed.), Mammals of Linchu, Shantung. Vertebrata PalAsiatica 12: 43–53. North America. Geochronology and , 153–210. Uni− Li, C.−K., Wu, W.−Y., and Qiu, Z.−D. 1984. Chinese Neogene: subdivision versity of California Press, Berkeley. and correlation. Vertebrata PalAsiatica 22: 163–178. Viret, J. and Zapfe, H. 1951. Sur quelques soricidés miocènes. Eclogae Martin, J.E. 1978. A new and unusual shrew (Soricidae) from the Miocene of geologicae Helvetiae 44: 411–426. Colorado and South Dakota. Journal of Paleontology 52 (3): 636–641. Wang, B.−Y. 1965. A new Miocene aceratherine rhinoceros of Shanwang, Martin, J.E. and Green, M. 1984. Insectivora, Sciuridae, and Cricetidae Shantung. Vertebrata PalAsiatica 9: 109–112. from the early Miocene Rosebud Formation in South Dakota. Special Wilson, R.W. 1960. Early Miocene rodents and insectivores from northeast− Publication of Carnegie Museum of Natural History 9: 28–40. ern Colorado. The University of Kansas Paleontological Contributions Mein, P. 1999. European Miocene mammal biochronology. In: G.E. Rössner, Vertebrata 7: 1–92. and K. Heissig (eds.), The Miocene Land Mammals of Europe, 25–38. Wilson, R.W. 1963. Notes on North American species of Heterosorex Verlag Dr. Friedrich Pfeil, München. Gaillard. Proceedings of the South Dakota Academy of Sciences 42: Qiu, Z.−D. 1981. A new sciuroptere from the middle Miocene of Linqu, 79–83. Shandong. Vertebrata PalAsiatica 19: 228–238. Xie, W.−M. 1979. First discovery of the Palaeotapirus in China. Vertebrata Qiu, Z.−D. 1996. Middle Miocene Micromamalian Fauna from Tunggur, PalAsiatica 17: 146–148. Nei Mongol. 216 pp. Science Press, Beijing. Yan, D.−F., Qiu, Z.−D., and Meng, Z.−Y. 1983. Middle Miocene Qiu, Z.−D. and Sun, B. 1988. New fossil micromammals from Shanwang, and fossil mammals from Shanwang, Shandong. Vertebrata PalAsiatica Shandong. Vertebrata PalAsiatica 26: 50–58. 21: 210–222. Qiu, Z.−X. and Qiu, Z.−D. 1995. Chronological sequence and subdivision of Young, C.C. 1937. On a Miocene mammalian fauna from Shantung. Bulle− Chinese Neogene mammalian faunas. Palaeogeography, Palaeoclimato− tin of Geological Society of China 17: 209–238. logy, Palaeoecology 116: 41–70. Young, C.C. 1977. Regarding the Anura and Chiroptera from Linqu, Qiu, Z.−X., Defa, D.−F., Jia, H., and Sun, B. 1986. The large−sized ursid fos− Shandong Province. Vertebrata PalAsiatica 15: 76–80. sils from Shanwang, Shandong. Vertebrata PalAsiatica 24: 191–206. Zhang, Y.−P. 1974. Miocene suids from Kaiynan, Yuannan and Linchu, Qiu, Z.−X., Yan, D.−F., Jia, H., and Sun, B. 1985. Preliminary observations Shantung. Vertebrata PalAsiatica 12: 117–123.

http://app.pan.pl/acta49/app49−357.pdf